Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “accelerator modeling”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 865 records · Page 48
All sourcesNASA NTRSDOE OSTI
DOE OSTI2024

Modeling Electric Vehicle Charging Load Using Origin-Destination Data

The accelerating adoption of electric vehicles (EVs) poses challenges to the power grid, necessitating precise representation of mobility patterns for effective infrastructure upgrades. Traditional simulation-based charging demand estimation faces limitations in generating trip chains reflective of actual travel patterns without complex network modeling. Hence, an innovative agent-based trip chain generation model is introduced to overcome these challenges. Drawing from the National Household Travel Survey (NHTS) and the NextGen NHTS origin-destination add-on data for Clarke County, Georgia, this study proposes a simulation method capturing both temporal and spatial mobility patterns without relying on extensive network topology data. The resulting trip chains predict EV charging load at the Census Block Group level, validated with a 1.03 correlation to actual trip counts, affirming their reflective accuracy. Two charging scenarios, residential-only and charging-everywhere, reveal distinct demand profiles. The charging-everywhere scenario aligns closely with the trip profile, while the residential-only scenario exhibits an afternoon peak slightly surpassing the former. This study contributes a data-driven charging demand estimation methodology, offering critical insights for grid resiliency planning amid the evolving landscape of EV adoption.

Pan, Melrose↗
DOE OSTI2021

A fast and accurate physics-informed neural network reduced order model with shallow masked autoencoder

Traditional linear subspace reduced order models (LS-ROMs) are able to accelerate physical simulations in which the intrinsic solution space falls into a subspace with a small dimension, i.e., the solution space has a small Kolmogorov n-width. However, for physical phenomena not of this type, e.g., any advection-dominated flow phenomena such as in traffic flow, atmospheric flows, and air flow over vehicles, a low-dimensional linear subspace poorly approximates the solution. To address cases such as these, we have developed a fast and accurate physics-informed neural network ROM, namely nonlinear manifold ROM (NM-ROM), which can better approximate high-fidelity model solutions with a smaller latent space dimension than the LS-ROMs. Our method takes advantage of the existing numerical methods that are used to solve the corresponding full order models. The efficiency is achieved by developing a hyper-reduction technique in the context of the NM-ROM. Numerical results show that neural networks can learn a more efficient latent space representation on advection-dominated data from 1D and 2D Burgers' equations. A speedup of up to 2.6 for 1D Burgers' and a speedup of 11.7 for 2D Burgers' equations are achieved with an appropriate treatment of the nonlinear terms through a hyper-reduction technique. Lastly, a posteriori error bounds for the NM-ROMs are derived that take account of the hyper-reduced operators.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2023

Final Technical Report: Innovative Magnetron Power Sources for Superconducting RF (SRF) Accelerators

In this project, MUONS is designing and building custom magnetrons at Richardson Electronics (RELL) to test the predictions of the models and make improvements to the designs to match the needs of NP SRF accelerators and their life-cycle costs. To increase lifetime, the filament temperature is controlled by a feedback circuit that includes a model of back streaming of electrons. The gain of the magnetron is optimized by the design of an input port. A permanent magnet design with trim coil for amplitude adjustment reduces energy consumption. Moving the construction of the magnetron from prototype to manufacturing includes the utilization of RELL processes and designs. To the extent that outgassing of materials influences the lifetime of the magnetron, methods for minimizing outgassing was explored. Further experimentation of the potential of sub-critical voltage operation continues to be explored. The commercialization goal of the project will be naturally achieved as RELL already has world-wide marketing, sales, and support to exploit the assembly-line capabilities generated by this project to enable magnetron technologies for NP accelerators.

43 PARTICLE ACCELERATORS↗
DOE OSTI2025

Giant Cloud Condensation Nuclei Facilitate Drizzle Formation in Stratocumulus—Insights From a Combined Observation‐Modeling Framework

The mechanism for initiating drizzle drop remains a gap in the current understanding of warm rain formation. One prevalent hypothesis suggests that the presence of Giant Cloud Condensation Nuclei (GCCN) generates drizzle‐sized drops necessary to trigger the Collision‐Coalescence (C‐C) process. Here, in this study, this hypothesis is investigated using a novel framework that integrates in situ observations, remote sensing measurements, and idealized models. Results show that GCCN can efficiently generate drizzle drops through condensation, producing a broad Droplet Size Distribution (DSD) comparable to in situ observations. The large drizzle drop and broad DSD strongly facilitate C‐C, further accelerating drizzle initiation. To compare with observation, the model‐generated DSDs are used to generate radar Doppler spectra where radar reflectivity and Doppler skewness is estimated. The simulated radar quantities correspond well with radar observations, providing critical evidence for the GCCN‐induced drizzle initiation mechanism.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2024

HDBind: encoding of molecular structure with hyperdimensional binary representations

Traditional methods for identifying “hit” molecules from a large collection of potential drug-like candidates rely on biophysical theory to compute approximations to the Gibbs free energy of the binding interaction between the drug and its protein target. These approaches have a significant limitation in that they require exceptional computing capabilities for even relatively small collections of molecules. Increasingly large and complex state-of-the-art deep learning approaches have gained popularity with the promise to improve the productivity of drug design, notorious for its numerous failures. However, as deep learning models increase in their size and complexity, their acceleration at the hardware level becomes more challenging. Hyperdimensional Computing (HDC) has recently gained attention in the computer hardware community due to its algorithmic simplicity relative to deep learning approaches. The HDC learning paradigm, which represents data with high-dimension binary vectors, allows the use of low-precision binary vector arithmetic to create models of the data that can be learned without the need for the gradient-based optimization required in many conventional machine learning and deep learning methods. This algorithmic simplicity allows for acceleration in hardware that has been previously demonstrated in a range of application areas (computer vision, bioinformatics, mass spectrometery, remote sensing, edge devices, etc.). To the best of our knowledge, our work is the first to consider HDC for the task of fast and efficient screening of modern drug-like compound libraries. We also propose the first HDC graph-based encoding methods for molecular data, demonstrating consistent and substantial improvement over previous work. We compare our approaches to alternative approaches on the well-studied MoleculeNet dataset and the recently proposed LIT-PCBA dataset derived from high quality PubChem assays. We demonstrate our methods on multiple target hardware platforms, including Graphics Processing Units (GPUs) and Field Programmable Gate Arrays (FPGAs), showing at least an order of magnitude improvement in energy efficiency versus even our smallest neural network baseline model with a single hidden layer. Our work thus motivates further investigation into molecular representation learning to develop ultra-efficient pre-screening tools. We make our code publicly available at https://github.com/LLNL/hdbind.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2020

Verification and Validation of High Explosive Reactive Burn Models Implemented in LANL's EAP and LAP Code Base

Reactive burn models represent a significant leap in high explosive (HE) modeling capability. The first generation of engineering models of HE detonation are called programmed burn models and they are largely based on the distance between a prescribed detonation point and each zone in a simulation. There have been many advancements to programmed burn models over the years and when the assumptions upon which they are based are met, a properly tuned programmed burn model can be highly accurate but if any of their assumptions is not met, as is the case for corner turning or weakly initiated HE burn, they will give the wrong answer. Reactive burn models represent an entirely new way of modeling HE burn. They use the local conditions of a zone – e.g. temperature, pressure or density – as calculated by a hydrocode to determine if and when the zone is going to detonate and if so, how rapidly. This difference opens up an entirely new set of capabilities for HE modeling. It makes it possible to accurately and predictively model phenomena like the effect of confinement and the formation of dead zones. Reactive burn models have seen sustained development effort at LANL for at least the last decade but several recent developments make it timely to transition reactive burn models from a research topic to a production tool. The main goal of this milestone is to facilitate and accelerate the adoption of reactive burn as a commonly available modeling option, with recommendations on the resolution that will be required and uncertainties associated with their modeling choices. To achieve this, we have performed verification, validation, and uncertainty quantification (UQ) assessments of AWSD and SURF/SURFplus in xRage and FLAG on a variety of different problems.

45 MILITARY TECHNOLOGY, WEAPONRY, AND NATIONAL DEF↗
DOE OSTI2022

Neural network accelerator for quantum control

Efficient quantum control is necessary for practical quantum computing implementations with current technologies. Conventional algorithms for determining optimal control parameters are computationally expensive, largely excluding them from use outside of the simulation. Existing hardware solutions structured as lookup tables are imprecise and costly. By designing a machine learning model to approximate the results of traditional tools, a more efficient method can be produced. Such a model can then be synthesized into a hardware accelerator for use in quantum systems. In this study, we demonstrate a machine learning algorithm for predicting optimal pulse parameters. This algorithm is lightweight enough to fit on a low-resource FPGA and perform inference with a latency of 175 ns and pipeline interval of 5 ns with > 0.99 gate fidelity. In the long term, such an accelerator could be used near quantum computing hardware where traditional computers cannot operate, enabling quantum control at a reasonable cost at low latencies without incurring large data bandwidths outside of the cryogenic environment.

43 PARTICLE ACCELERATORS↗
DOE OSTI2020

Characteristics of the Diffuse Astrophysical Electron and Tau Neutrino Flux with Six Years of IceCube High Energy Cascade Data

In this work, we report on the first measurement of the astrophysical neutrino flux using particle showers (cascades) in IceCube data from 2010-2015. Assuming standard oscillations, the astrophysical neutrinos in this dedicated cascade sample are dominated (~90%) by electron and tau flavors. The flux, observed in the sensitive energy range from 16 TeV to 2.6 PeV, is consistent with a single power-law model as expected from Fermi-type acceleration of high energy particles at astrophysical sources. We find the flux spectral index to be γ=2.53±0.07 and a flux normalization for each neutrino flavor of $\phi$ astro =1.66$_{-0.27}^{+0.25}$ at E 0 =100 TeV, in agreement with IceCube's complementary muon neutrino results and with all-neutrino flavor fit results. In the measured energy range we reject spectral indices γ≤2.28 at ≥3σ significance level. Because of high neutrino energy resolution and low atmospheric neutrino backgrounds, this analysis provides the most detailed characterization of the neutrino flux at energies below ~100 TeV compared to previous IceCube results. Results from fits assuming more complex neutrino flux models suggest a flux softening at high energies and a flux hardening at low energies (p value ≥0.06). The sizable and smooth flux measured below ~100 TeV remains a puzzle. In order to not violate the isotropic diffuse gamma-ray background as measured by the Fermi Large Area Telescope, it suggests the existence of astrophysical neutrino sources characterized by dense environments which are opaque to gamma rays.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗
DOE OSTI2020

Self-accelerated corrosion of nuclear waste forms at material interfaces

The US plan for high-level nuclear waste includes the immobilization of long-lived radionuclides in glass or ceramic waste forms in stainless-steel canisters for disposal in deep geological repositories. Here we report that, under simulated repository conditions, corrosion could be significantly accelerated at the interfaces of different barrier materials, which has not been considered in the current safety and performance assessment models. Severe localized corrosion was found at the interfaces between stainless steel and a model nuclear waste glass and between stainless steel and a ceramic waste form. The accelerated corrosion can be attributed to changes of solution chemistry and local acidity/alkalinity within a confined space, which significantly alter the corrosion of both the waste-form materials and the metallic canisters. The corrosion that is accelerated by the interface interaction between dissimilar materials could profoundly impact the service life of the nuclear waste packages, which, therefore, should be carefully considered when evaluating the performance of waste forms and their packages. Moreover, compatible barriers should be selected to further optimize the performance of the geological repository system.

Guo, Xiaolei↗
DOE OSTI2020

HERON

HERON is a modeling toolset and plugin for RAVEN to accelerate stochastic technoeconomic assessment of the economic viability of various grid-energy system configurations, especially with application to electrical grids and integrated energy systems (IES). The frameworks enabling this toolset are INL's RAVEN and its TEAL plugin. The toolset includes automated RAVEN workflow generation tools for performing stochastic analysis and optimization driven by technoeconomic factors. The workflows generated employ RAVEN for parametric or optimizing study of grid component characteristics, and again for stochastic sampling of regulated and deregulated markets, load, and weather histories. The toolset further provides methods for economically-optimal dispatch of system components given the stochastic samples of market, load, and weather histories. The toolset allows assembling of existing and new models through the RAVEN interface to the Modelica process modeling language.

Talbot, PaulW↗
DOE OSTI2025

2025 Status Report - Fiberboard Properties and Degradation Rates for Storage of 9975 Shipping Package in KAC

Fiberboard assemblies in the 9975 package were identified as susceptible to aging and degradation under long-term storage conditions. Thermal, mechanical, and physical properties have been measured on fiberboard samples aging for up to 20 years at accelerated conditions. The aging environments include elevated temperature up to 250 ºF (the maximum allowed service temperature for fiberboard in 9975 packages) and elevated humidity. Accelerated aging results have been analyzed and used to build aging models. Correlations relating several properties (thermal conductivity, energy absorption, weight, dimensions, and density) to their rate of change in potential storage environments have been developed. Combined with an estimate of the actual conditions the fiberboard experiences in K-Area Complex (KAC), these models allow development of service life predictions. KAC completed calculations that supported extending the service life of 9975 packages with 3013 containers in storage from 20 years to 40 years. These calculations addressed the potential for degraded fiberboard properties following 40 years in storage using degradation models. The 40 year storage evaluations assumed the degradation rates from initial values for density, dimensional, and thermal properties were 0.5% per year (i.e., after 40 years, the density, dimensional, and thermal properties of fiberboard were 80% of its initial values). The updated degradation models developed in this report were compared to the assumptions in those calculations, and the current results for fiberboard dimensions, density, axial and radial thermal conductivities, and energy absorption remain consistent with those calculations. Additional data will continue to be collected to permit future refinements to the models and assumptions. Some of the predicted degradation rates presented in this report are purposely high to accelerate aging. These rates relate to environments that do not exist within KAC or are postulated to only occur as upset conditions that are unlikely to persist for an extended period. For a typical package stored in KAC with approximately 10-12 watts internal heat load or less, and ambient temperatures below 90 ºF, the fiberboard assemblies experience storage conditions less severe than any of the aging environments (i.e., under 125 °F) used in this study. Fiberboard in conforming packages with lower internal heat loads should experience little or no degradation and is expected to provide a service life beyond the currently approved 40 year storage period. Packages with higher internal heat loads may not continue to perform their required safety functions beyond 40 years. Ultimately, service life will be determined by the cumulative effect of degradation from all the conditions these packages might encounter. The results and model predictions presented in this report are applicable to 9975 packages with cane or softwood fiberboard overpack assemblies. These degradation models do not address the effects of nonconforming conditions such as the presence of excess moisture and mold or beetle infestations.

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗
DOE OSTI2024

FiberFlex: Real-time FPGA-based Intelligent and Distributed Fiber Sensor System for Pedestrian Recognition

In recent years, security monitoring of public places and critical infrastructure has heavily relied on the widespread use of cameras, raising concerns about personal privacy violations. To balance the need for effective security monitoring with the protection of personal privacy, we explore the potential of optical fiber sensors for this application. This article proposes FiberFlex, an intelligent and distributed fiber sensor system. Ultizing Field Programmable Gate Arrays (FPGA) high-level synthesis (HLS) acceleration, FiberFlex offers real-time pedestrian detection by co-designing the entire pipeline of optical signal acquisition, processing, and recognition networks based on the principles of optical fiber sensing. As a promising alternative to traditional camera-based monitoring systems, FiberFlex achieves pedestrian detection by analyzing the vibration patterns caused by pedestrian footsteps, enabling security monitoring while preserving individual privacy. FiberFlex comprises three modules: First , fiber-optic sensing system: A fiber-optic distributed acoustic sensing (DAS) system is built and used to measure the ground vibration waves generated by people walking. Second , algorithms: We first collect the training data by measuring the ground vibration waves, label the data, and use the data to train the neural network models to perform pedestrian recognition. Third , hardware accelerators: We use HLS tools to design hardware modules on FPGA for data collection and pre-processing and integrate them with the downstream neural network accelerators to perform in-line real-time pedestrian detection. The final detection results are sent back from FPGA to the host CPU. We implement our system FiberFlex with the in-house built DAS system and AMD/Xilinx Kintex7 FPGA KC705 board and verify the whole system using the real-world collected data. We conduct recognition tests on five test subjects of varying ages, heights, and weights in a fixed sensing area. Each subject experienced 20 real-time recognition tests using their daily walking habits, and the subjects were given adequate rest between tests. After 100 tests on five test subjects, the overall real-time recognition accuracy exceeded \(88.0\%\) . The whole system uses 55 W of power, 33 W in the optical DAS system and 22 W in the FPGA. Relying on its end-to-end interdisciplinary design, FiberFlex seamlessly combines fiber-optic sensors with FPGA accelerators to enable low-power real-time security monitoring without compromising privacy, making it a valuable addition to the existing security monitoring network. According to FiberFlex, more valuable research can be conducted in the future, such as fall monitoring for the elderly, migration of identification networks between different application scenarios, and improvement of anti-interference performance in more complex environments. In future perception networks, where the “eyes” are not feasible, let’s use fiber optic touch instead.

Distributed↗
DOE OSTI2021

A Survey: Handling Irregularities in Neural Network Acceleration with FPGAs

In the last decade, Artificial Intelligence (AI) through Deep Neural Networks (DNNs) has penetrated virtually every aspect of science, technology, and business. Many types of DNNs have been and continue to be developed, including Convolutional Neural Networks (CNNs), Recurrent Neural Net- works (RNNs), and Graph Neural Networks (GNNs). The overall problem for all of these Neural Networks (NNs) is that their target applications generally pose stringent constraints on latency and throughput, while also having strict accuracy requirements. There have been many previous efforts in creating hardware to accelerate NNs. The problem designers face is that optimal NN models typically have significant irregularities, making them hardware-unfriendly. In this paper, we first define the problems in NN acceleration by characterizing common irregularities in NN processing into 4 types; then we summarize the existing works that handle the four types of irregularities efficiently using hardware, especially FPGAs; finally, we provide a new vision of next-generation FPGA-based NN acceleration: that the emerging heterogeneity in the next-generation FPGAs is the key to achieving higher performance.

Geng, Tong↗
DOE OSTI2024

Infrastructure improvements in the National Reactor Innovation Center Virtual Test Bed

The National Reactor Innovation Center’s (NRIC) mission is to support deployment of novel reactor concepts. This is achieved by providing physical and virtual spaces for building and testing various components, systems, and complete pilot plants. The Virtual Test Bed (VTB) represents the virtual counterpart to the physical test bed. The VTB is being developed in collaboration with the Department of Energy’s (DOE) Nuclear Energy Advanced Modeling and Simulation (NEAMS) program. The mission of the VTB is to accelerate the deployment and licensing of advanced reactors by leveraging state-of-the-art modeling and simulation (M&S) tools developed by the DOE NEAMS program. This is accomplished via three primary means: (1) openly hosting simulations that showcase analysis capabilities, (2) continuously testing the models hosted against code updates to avoid deprecation, and (3) filling key M&S gaps that are relevant for the physical NRIC test beds. The VTB repository consists of two sub-entities: • A documentation website detailing the models: https://mooseframework.inl.gov/virtual_test_bed • A GitHub repository that hosts the corresponding files: https://github.com/idaholab/virtual_test_bed. This paper presents the infrastructure added to the VTB in the last two fiscal years. Additional information about the VTB can be found in various publications [1, 2, 3, 4, 5, 6, 7, 8].

22 - GENERAL STUDIES OF NUCLEAR REACTORS↗
DOE OSTI2024

Status of New Models Hosted on the Virtual Test Bed (VTB) in 2024

The National Reactor Innovation Center (NRIC) mission is to support deployment of novel reactor concepts. This is achieved by providing physical and virtual spaces for building and testing various components, systems, and complete pilot plants. The Virtual Test Bed (VTB) represents the virtual counterpart to the physical test bed. It is in development in collaboration with the Department of Energy’s (DOE) Nuclear Energy Advanced Modeling and Simulation (NEAMS) program. The mission of the VTB is to accelerate the deployment and licensing of advanced reactors by leveraging state-of-the-art modeling and simulation (M&S) tools developed by the DOE NEAMS program. This is accomplished by three primary means: (1) openly hosting simulations that showcase analysis capabilities, (2) continuously testing the models hosted against code updates to avoid deprecation, and (3) filling key M&S gaps that are relevant for the physical NRIC test beds. The VTB repository consists of two sub-entities: 1. A documentation website detailing the models (https://mooseframework.inl.gov/virtual_test_bed). 2. A GitHub repository that hosts the corresponding files (https://github.com/idaholab/virtual_test_bed). Previous documentation on the models hosted in the VTB can be found in [1,2,3,4]. These references also include additional background information on the various NEAMS codes showcased in the VTB (which is omitted here for brevity). This paper primarily provides a status update of the most recent additions to the repository.

22 - GENERAL STUDIES OF NUCLEAR REACTORS↗
DOE OSTI2023

Achieving Higher Fidelity Pulsed-Power Simulations Using Advanced Gap Closure Modeling

This project has progressed several physics models in the EMPIRE plasma simulation code to achieve higher fidelity simulations of high-current diode operation in pulsed-power accelerators. In this report, we present details for the following major work products: (1) a set of verification problems covering all key processes involved in gap closure physics has been designed; this suite has facilitated feature vetting and overall model maturation, (2) a new EMPIRE exemplar has been developed: the Radiographic Integrated Test Stand 6 (RITS-6) diode, and (3) An exemplar for the Saturn accelerator exemplar was enabled by the models matured under this work to self-consistently simulate further into the diode pulse than previously possible (bipolar flow regime). These developments have lead to the highest confidence EMPIRE power flow predictions of the Saturn accelerator to date. Additionally, three modeling approaches for simulating electrode plasmas have been investigated. We report on these results and provide recommendations.

43 PARTICLE ACCELERATORS↗
DOE OSTI2020

Modeling synergistic fuel cell membrane degradation with mitigating effects of cerium

During operation, polymer-electrolyte fuel cells undergo mechanical and chemical degradation, which behave synergistically and lead to accelerated membrane deterioration over time. This study builds upon previous modeling work on mechanical degradation, as described by a pinhole in the membrane, and chemical degradation mitigation by using cerium. By combining these two models, analysis can be carried out on the coupled degradation methods and how the mitigation effects of cerium disrupt the degradation cycle. The model results show how the presence of a pinhole in the membrane changes the distribution of cerium in the cell and modifies chemical degradation rates. In addition, the model shows how cerium affects the rate of pinhole growth by slowing down the rate of change in mechanical properties due to chemical degradation. Finally, the model shows how cerium modifies the mechanical and chemical degradation rates of the membrane under humidity- and voltage-cycling conditions.

Ehlinger, VM↗
DOE OSTI2022

Fixing Amdahl's Law within the Limits of Accelerated Systems: FALLACY

Closeout report for FALLACY project. The performance of Data Model Convergence Initiative (DMC) applications on parallel machines is far below the limit set by Amdahl’s law. Whether the machine is based on many-core, GPUs, FPGAs, or a heterogeneous combination, usually the most significant bottleneck is accessing data from the memory system. Aligning with DMC’s HW/architecture thrust, this project developed a set of memory-centric tools called ‘MemGaze’ that inform the HW/SW stack about an application’s memory behavior, including data access latency and diagnosing poor data layout and data composition. Our approach uses architectural modeling and analysis of workload data accesses.

97 MATHEMATICS AND COMPUTING↗